Brownouts, shorts, explosions and the unix pc.

[Thad P Floryan]

dt at yenta.alb.nm.us (David B. Thomas)
in <1991Jan9.064342.7045 at yenta.alb.nm.us> writes (re: Tesla coils):

	So how big, physically, would this destructive device be?  And
	would you simply zap it in the vicinity, and the massive EMI
	will do in sensitive equipment?  Are our military's electronic
	items susceptible as well?

I haven't studied Tesla coils all that much, but the dealer from whom I
bought my surge protectors and UPS systems has a neato mobile display which
he takes to customer sites and part of the display is a Tesla coil that
weights about 40 pounds and is about the size of a small kid ... it has been
known to zap IBM-PCs in adjacent rooms (no great loss, right! :-)  He has
stopped using that Tesla coil due to requests from his insurance company.

A big concern should there ever be a nuclear war is that the EMP would take
out the entire telephone system, all computers, all radios, all radar, traffic
lights, etc etc etc so my answer would be "Yes, our military's electronic
items are susceptible as well."

And do NOT bring up "Tempest" shielding;  that's only to prevent RFI emissions
FROM military equipment being detected, intercepted and decoded.  One of the
original requirements for DES was simply to "protect" data travelling on wires
between a CPU and its peripherals ... those RFI emissions CAN be detected and
sensitive info so broadcast can be picked up and used by spies and competitors.
Even the RFI hash from the front of a CRT can be relatively-easily "decoded"
and an exact image of a CRT, say, 100 feet away, can be reconstructed on
another CRT; in other words, your sensitive spreadsheet info on your screen
in a closed room can be "intercepted" (by picking up the RFI) and displayed on
another CRT.

Operating a computer in something like a Faraday Cage may serve to reduce the
effects of the "problems" mentioned above.  I remember back when I worked for
the Electronic Defense Labs how we HAD to use such shielded rooms just to be
ablse to measure some of the things we were doing ... other equipment in the
lab ('scopes, voltmeters, etc.) would otherwise interfere our measurements
(and this was back in the early '60's).

I would venture to speculate that all such problems can be solved if one is
willing to spend enough money and endure the concomitant inconvenience(s).

Sorta like the goal of "security" is to make it cost more in time and effort
than the breach is worth, one has to make a decision as to how much the peace
of mind is worth; and locks can only keep out "honest" people, not a thief! :-)

Thad Floryan [ thad at cup.portal.com ]